welcome: please sign in
location: simple


A Simple Radio

by Glen Gardner


A few months back I made the comment that it ought to be possible to build a simple radio that had good performance and is usable as a communications receiver, with off the shelf pre-assembled modules as building blocks.

Recently, I decided to act on that idea and began gathering parts to build a simple HF receiver. I wanted good sensitivity, and I wanted all band coverage.

I decided on a basic direct-conversion double-sideband receiver. I could use readily-available ring mixers for the product detector, and an inexpensive Chinese-made DDS synthesizer for the VFO. For the AF preamp, an inexpensive high gain microphone preamplifier module can be used.



The DDS board was ordered on E-Bay for $29 plus shipping. It was advertised as a 50MHz AD9851 DDS board with sine and square wave output and LCD display. The web site claimed the output was "about a volt". The DDS board came with a 9V switching-type wall-wart power supply, and a test lead with BNC and alligator clip terminations. There was NO documentation.

Some things to fix

When I got the DDS board I powered it up and found that the sine wave output was unacceptably low at 200 mV @ 7MHz. The output at 50 MHz was just a few mV. The datasheet for the AD9851 specifies that the output is approximately 500 mV. There were a large number of low amplitude (-50 dB) spurs in the spectral plot as viewed with a digital sampling oscilloscope. The square wave output was erratic and did not work above around 10 MHz.

Noisy stuff

On close examination, it was found that the switch (marked "CH") on the board that is used to select square/sine output, coupled some of the square wave to the output when the switch was set to select sine wave. Removing the switch and replacing it with a jumper to permanently select sine wave output resulted in a significant reduction in spurious output.

More Oddities

For whatever reason, the designer had placed a 49 ohm series resistor (R15) in series with the output of the DDS, causing the output to drop by half when a 50 ohm load was connected. Replacing the resistor with a 0.1uF size 0805 SMT capacitor resulted in about 450 mV output into a 50 ohm load, close to the correct level as specified by Analog Devices. The resistor R15 is located on the top side of the board between the sine/square wave switch, CH and the BNC connector.

Having done all of these things I ended up with a pretty good DDS board. Sweeping across the 1Hz-50MHz frequency tuning range of the DDS, I discovered, that after the modifications, the cut-off frequency of the on-board filter on the DDS was exactly 30 MHz. The filter was designed for a 30 MHz cut-off frequency(??!!). The on-board filter does work well, and after fixes, there were no problems with the mirror frequency, and no significant spurious products were observed.

NOTE: As of this writing , I have had the dds boards a week. Now you can go to E-Bay and see the same vendors selling what appears to be the same board, with the same specifications, but if you look at the pictures , they now show a board with a slightly different layout. I suspect that you basically get whatever they have. There were some 30 MHz versions of these boards around. I suspect that maybe some vendors have been reprogramming the controller to cover 50 MHz without upgrading the filter on the board. Some boards may actually have 50 MHz filters in them, but the ones which I received most definitely start rolling off at 30 MHz. I bought four of these boards and while they vary a little bit due to tolerances, they are all very much the same.... Caveat Emptor... In this case the boards can be made to work for this particular purpose, so all is well.

The Ring Mixer

I opted to use a Minicircuits ZP3+ ring mixer. It has better low-end frequency response and lower conversion loss as compared to ZP-1 and SBL-1. It costs about $40 and is supplied in a machined aluminium case with BNC connectors. A similar mixer that will also work well is the Minicircuits ZP-1 (also at about $40). For those on a budget, and don't mind wiring non-connectorized components, there is the Minicircuits SBL-1 for $7. It has specifications similar to the ZP-1 but is a through-hole component.

The RF Preamplifier

The noise figure and sensitivity of the receiver are largely limited by the gain and noise figure of the RF preamplifier. This motivated the selection of a pretty good preamplifier instead of a cheap one. The preamplifier is made by Advanced Receiver Research (AR^2). Model P1-30/20VD. It sells for about $46 plus shipping. It covers 1 MHz to 30 MHz with 20 dB gain and a noise figure of 2.5 dB or better.

A cheaper alternative to this might be one of the MMIC preamplifiers that are made by a number of manufacturers. One example is the preamplifier kit offered by TenTec, the T-kit No 1001 Broadband RF Preamplifier kit. It produces 18.5 dB gain from 100 kHz-1000 MHz. It has a noise figure of 5.5 dB and is unconditionally stable. It is easy to build and sells for about $13 plus shipping. The MMIC preamps will work rather well, but not quite as nicely as the AR^2 preamp above. One advantage of using the MMIC preamplifier is that you get coverage down to 100 KHz! The ZP-3+ mixer is rated for use down to 150KHz, but should be usable down to 100 KHz.

The Low-pass Filter

I used a Minicircuits BLP-30+ lowpass filter. This is a very compact, connectorized lowpass filter that exhibits a sharp rolloff at 32 MHz. Similar filters are readily made using discrete components , with Dan's kits and parts offering a variety of bandpass filters as well.

The purpose of this filter is to help reduce unwanted responses to signals outside the passband (1-30 MHz). It costs about $35. Alternatively, you can make your own low-pass filter, or you could even make bandpass filters for your favourite bands. Be sure to put the filter AFTER the preamplifier, not between it and the antenna to avoid degrading the noise figure of the radio.

NOTE: I use a magnetic loop antenna, and it has very selective tuning that does a good job of rejecting most out of band signals. Because of this I really don't need much for filtering ahead of the RF stage as the antenna acts like a preselector. If you use an antenna with wider bandwidth (like a dipole), and live in an area with lots of strong signals, you may experience some problems caused by overload from strong out of band signals. In that situation I would recommend either using a magnetic loop antenna, or adding a preselector, or bandpass filters between the radio and antenna. This little radio has excellent dynamic range so this won't be a problem for most people.

The AF Preamplifier

I used a low-cost dynamic microphone preamplifier in kit form from Future Kit, the Model FK647 Dynamic Microphone Preamplifier. It costs about $3. This is a simple kit, taking me about 10 minutes to assemble. I was rather surprised by its performance. The gain is very high, and it makes very little noise. The input impedance is high. You will need to connect a 0.047 uF capacitor across the input terminals to decouple it from the mixer to avoid excessive amounts of hum in the output. The output appears to be close to line level, and suitable for connecting to an audio amplifier or computer sound card. The frequency response of the preamplifier is very flat, up to many, many kilohertz, so the output needs to go to some kind of filter or tone control to limit the AF bandwidth. It is a good idea to provide this board with its own voltage regulator to isolate it from power supply noises. I used a 9.1V 1W Zener diode tacked onto the supply rails, on the underside of the board, with a 180 ohm series dropping resistor connected to the 12 VDC supply.

Filtering the AF output

I connected the output of the AF preamp to my switched capacitor audio filter (also on this site), and fed the output to a volume control and a 3W audio amplifier. It sounded quite nice. Without the filter, it sounds quite harsh, but the radio is still very usable. A simple unity gain, second order lowpass audio filter with a cutoff at 1.8 kHz would probably work nicely. Future Kit makes a nice active tone/volume control unit that may work well for this (I have not tried it). It is available assembled or in kit form for about $7. Future Kit FK625 tone control mono

The AF Amplifier

You can use most generic audio amplifier kits with this. You could also plug the AF preamp into a pc sound card. I used a 3W audio amplifier which I built from a TDA820 a long time ago.

Putting it together


This is very simple. I attached everything to a large aluminum metal plate that i had left over from an old project. I mounted the mixer to the plate using double-sided foam tape. I used the plate as the common grounding point for everything. Make sure everything that has a ground is grounded to the plate.

1) The rf preamp output goes to the 30 MHz lowpass filter.

2) The other end of the 30 MHz lowpass filter goes to the RF input of the mixer.

3) The RF output from the DDS board goes to the LO input on the mixer.

4) The IF out of the mixer (often marked "X" on the mixer) goes to the AF preamp input.

5) The AF output of the AF preamplifier goes to a pc sound card, or to an audio filter, volume control, amp... as discussed above.

Hooking Up Power

The DDS board requires 7-9VDC at a fair amount of current. It comes with a wall wart that is prone to causing electrical noise when and where you don't want it to. Don't use it. Instead get an LM7808 or an MC7808, 8V 1A TO-220 package regulator IC and use the output of that to power the DDS. You should bolt the regulator heat sink to the metal chassis plate to provide the necessary ground connection and to provide a convenient heat sink. I had a scrap board with a heat sink and an 8V, 1A regulator on it, so I used it as the regulator for the DDS.

The AR^2 RF preamplifier will operate directly from the 12VDC supply.

The AF preamp needs 9 VDC, and should have its own regulator, or it may become noisy. I used a 9.1V 1W Zener diode along with a 180 ohm series dropping resistor to regulate the 12VDC supply down to 9V. Alternatively, using an LM78L08 voltage regulator in a TO-92 package will suffice.

You want to power all of this stuff with a clean, regulated source of power. It should be okay to operate this radio from a 11-15VDC power supply capable of delivering about 600 mA.

simple (last edited 2014-08-02 01:08:07 by GlenGardner)